Note : Les descriptions sont présentées dans la langue officielle dans laquelle elles ont été soumises.
CA 03031038 2019-01-16
1
METHOD AND DEVICE FOR INSPECTING THE DAMAGE ON THE SKIN OF AN
AIRCRAFT AFTER A LIGHTNING STRIKE
FIELD OF THE INVENTION
The present invention refers to a method and a device for aircraft damage
inspection after a
lightning strike by getting an optimized admissible visual damage determined
by the
relationship between the structural damage and the visual damage produced by a
certain
lightning strike energy.
BACKGROUND OF THE INVENTION
Lightning strikes can affect airline operations causing costly delays and
service interruptions.
When commercial aircraft are struck by lightning, the result can range from no
damage to
severe damage which may require immediate repair that can take the aircraft
out of service
for an extended period of time. The severity of the damage varies greatly and
is dependent
on multiple factors such as the energy level of the lightning strike.
Lightning initially attaches to an aircraft extremity at one spot and exits
from another. Due to
the relative speed of the aircraft to the lightning channel, there may be
several additional
attachment points between those initial and exit points (swept stroke), this
may lead to up to
hundreds of marks spread out all along the skin of the aircraft.
The direct effects of a lightning strike are the damages caused to the
structure which include
melt through, pitting to the structure, burn marks around fasteners and burnt
paint for metallic
structures. In the case of composite structures, apart from burnt paint,
damaged fibers and
delaminations can be found.
At each attachment point of the lightning arc, the damage on a composite
component can be
separated into two parts:
- Firstly, at the exterior surface of the fuselage burn marks and missing
paint are
visible and quantifiable.
CA 03031038 2019-01-16
2
- Secondly, inside the composite material, structural damages as
overheating of the
resin, broken fibers and delamination can be found.
Some zones of the aircraft are more prone to lightning strikes than others. It
is already known
and standardized that the aircraft can be divided into several zones regarding
the probability
.. of receiving a lightning strike:
Zone 1 - High probability of initial lightning strike attachment (entry or
exit).
Zone 2 - High probability of a lightning strike being swept from a point of
initial
attachment.
Zone 3 - Any aircraft surface other than those covered by zones 1 and 2.
At each zone, a maximum energy level of the strike is expected according to
existing
standards, for instance, Eurocae ED-91 & ED-84.
When the aircraft is back on the ground after being struck, the severity of
the damage shall
be assessed and, depending on the size of the damages, some maintenance
operations
(repairs, inspection of areas, etc.) can be necessary.
In CFRP (Carbon-fiber¨reinforced polymer) structures, it is well known that
delamination
damages can be greater than the external observed damage (the visual damage),
reason
why, with the current state of the art, an inspection with a device capable of
detecting
damage within the structure is required; typically these test devices use
ultrasound, X-ray or
thermography. The inspection is performed mark by mark spending between five
to ten
minutes at each mark; therefore, in a case when there are substantial
lightning attachment
points (swept stroke), the time associated to those inspections at each mark
may require the
aircraft to be on ground for a significant time, putting in risk the daily
operations of the
airline/air carrier. Thus the inspection of the aircraft after a lightning
strike is time consuming
and can only be done by qualified operators.
CA 03031038 2019-01-16
3
SUMMARY OF THE INVENTION
The object of the invention is a method and a device for aircraft damage
inspection after
impact of a lightning strike based on an optimized admissible visual damage
threshold.
After being struck by lightning, the CFRP skin comprises a set of visual
marks. Some of
them have both paint removed off and potential structural damage.
The inspection method, object of the invention, is based on the link of these
two
parameters:
- The Visual Damage (VD), wich refers to the area wherein the paint has
been
scrapped off the CFRP skin by the lightning strike.
- The Structural Damage (SD) resulting from the lightning strike which
refers to the
damage on structural layers of the composite elements wherein laminate is
effectively
damaged.
A deep study on visual damage versus structural damage has been performed in
the full
range of both parameters in order to validate an aircraft release policy based
on confidence
of the predicted results.
This defined relationship enables the definition of the structural damage that
can be
expected for a given visual damage. The acceptance criteria (visual damage)
must be
defined in such a way that the impacted aircraft parts are able to sustain the
level of
structural loads that allow the continuous safe operation of the aircraft.
The method of inspection object of the invention allows determining a maximum
visual
damage for which it is assured that the structural damages associated to it
are always
below the defined limits. This admissible visual damage is referred as VDADL
from now on.
Relationship between visual damage and structural damage
The relationship between visual damage and structural damage has not been
studied either
by the industry or the scientific community.
CA 03031038 2019-01-16
4
By choosing the right main parameter that drives the relationship and making
some
hypothesis, the invention develops a method to predict one of the parameters
(VD or SD),
when the other is given, with a defined confidence level. The conclusions of
this method
have been verified in a laboratory environment and in real flight conditions.
Primary parameter. ¨ Several parameters have an impact on the structural
damage such
as paint thickness, CFRP thickness, percentage of dielectric thickness of the
CFRP,
structural reinforcement, the peak current of the lightning, the energy
injected by the
lightning, or what is the same, the energy level of the lightning strike, etc.
The parameter
chosen as "primary" or "main" parameter has a critical importance, as it
allows to find a
dependency. Not all of the previous parameters have an influence or the same
level of
influence on the visual damage. In this case, after analyzing multiple
parameters, the
energy level of the lightning strike or the peak current of the lightning
strike is chosen as the
primary parameter.
Analyzing the empirical data, the existence of a monotonically increasing
relationship
between the average visual damage and the energy level of the lightning strike
or the peak
current, as well as between the average structural damage and the energy level
or the peak
current, is observed. This observation will be the base for establishing
different conclusions
for this method.
Assumption. ¨ It is considered that the rest of aforementioned parameters
(e.g. paint
thickness, CFRP thickness, structural reinforcements) are less significant in
comparison
with the main parameter, assuming these parameters are within the threshold
defined by
the aircraft manufacturer (e.g. paint thickness may be limited by
manufacturing processes,
operational requirements and expectable damages). Therefore, these secondary
parameters can be absorbed as part of the natural randomness that
characterizes the
lightning phenomenon.
An admissible Structural Damage (ADL) is provided for every component of the
aircraft, i.e.
for the aircraft zone in which the mark is located. Said admissible Structural
Damage (ADL)
is set according to the technical features of the material of the component
and the structural
loads the aircraft component will see during its lifetime.
The method object of the invention comprises the following steps:
CA 03031038 2019-01-16
- in every mark, measuring the area of paint that has been removed by the
lightning strike (the Visual Damage),
- comparing the measured area with an area threshold value, wherein the
threshold value is related to the Structural Damage (SD) of every mark and is
obtained by the following steps:
o sorting empirical data of Structural Damage (SD) versus Visual Damage
(VD) for different levels of energy of the lightning strike, the levels of
energy being below a maximum energy expectable in the specific aircraft
zone in which the mark is located,
o providing an admissible Structural Damage (ADL) for the aircraft zone in
which the mark is located,
o calculating an admissible Visual Damage (VDADL) for which the maximum
Structural Damage (SD) that can be encountered is less than the
admissible structural damage (ADL) for the aircraft zone in which the
mark is located, said optimized Visual Damage (VDADL) being the area
threshold value.
Note the reduced complexity of the proposed method compared to the state of
the art
method. Time-consuming non-destructive tests are avoided; the operator just
has to
measure the visual damage caused by the lightning strike and compare it to an
already
calculated threshold value.
After choosing the primary parameter and setting the assumption, the following
embodiments are developed to establish the method for calculating the
admissible Visual
Damage (VDADL):
- if the most energetic lightning strike expected in the specific aircraft
zone in
which the mark is located causes a Structural Damage (SD) lower than the
provided admissible Structural Damage (ADL), a method called "Bounding-Box
Approach" is used,
A
CA 03031038 2019-01-16
6
-
if the most energetic lightning strike expected in the specific aircraft
zone causes
a Structural Damage (SD) higher than the provided admissible Structural
Damage (ADL), a method called "Energy-Dependent Approach" is used.
Both methods determine an admissible Visual Damage (VDADL) for which the
maximum
Structural Damage (SD) that can be encountered is less than the Admissible
Damage (ADL)
for the skin zone in which the mark is located. Therefore, if damages caused
by lightning
strike are found during an inspection and the Visual Damage (VD) is smaller
than
admissible Visual Damage (VDADL), it is possible to ensure the continuous
operation of the
aircraft without performing any non-destructive test, saving time for the
airline.
It is also an object of the present invention a device for inspecting the
damage on the
aircraft after a lightning strike. It comprises a characteristic area that is
equal to a Visual
Damage threshold value which is related to the Structural Damage and wherein
the
threshold value is calculated according to the previous method.
Said device allows a quick comparison of the Visual Damage of every mark by
just locating
the device over the mark. The advantage is that it provides even a quicker
visual inspection
that leads to a reduction of the time on ground of the aircraft.
With this method, most of the damages, wherein an inspection with an apparatus
of
ultrasounds, X-ray or thermography was needed, are now assessed only by visual
inspection, saving hours and increasing the operability of the aircraft.
DESCRIPTION OF THE FIGURE
To complete the description and in order to provide for a better understanding
of the
invention, drawings are provided. Said drawings are an integral part of the
description and
illustrate preferred embodiments of the invention. The drawings comprise the
following
figures.
Figure 1 shows graphics depicting the behavior of the Structural Damage (SD)
and Visual
Damage (VD) as the energy level of a lightning strike increases. Figure la
represents the
CA 03031038 2019-01-16
7
case when the method called "Bounding-Box Approach" must be applied (maximum
Structural Damage (SD) expected in that aircraft zone is less than the
admissible
Structural Damage (ADL)). Figure lb represents the alternative case, when the
method
called "Energy-Dependent Approach" must be used (maximum Structural Damage
(SD)
expected in that aircraft zone is greater than the admissible Structural
Damage (ADL)).
Figure 2 shows the two different approaches that can be used to find the
Visual Damage-
Structural Damage (VD-SD) relationship. The rectangles represent the "Bounding-
Box
Approach" while the other increasing lines represent the "Energy-Dependent
Approach".
"Energy level 1" is the least energetic and "energy level 4" is the most
energetic.
Figure 3 shows a qualitative representation of the "Bounding-Box Approach".
Figure 4 shows the results obtained with the "Bounding-Box Approach" for a
real example.
Figure 5 shows a graphic depicting a comparison of qualitative results when
applying the
"Bounding-Box Approach" and the "Energy-Dependent Approach".
Figure 6 shows a graphic depicting the nature of the Visual Damage-Structural
Damage
(VD-SD) relationship and compares it to the statistical method used.
Figure 7a shows graphics depicting the variation of Structural Damage (SD)
prediction
intervals with respect to the maximum energy level tested, being ei the
highest energy
level expected on a specific zone of the aircraft in which the mark is located
and ea the
lowest one. Figure 7b highlights the possible problem of obtaining structural
damages on
areas which initially may thought to be safe due to the maximum energy level
chosen.
Figure 8 shows a graphic depicting the energy level for calculating the
admissible Visual
Damage (VDADL), and the minimum and maximum Visual Damages (VDmin, VDmax) for
said
energy level.
Figures 9a and 9b show graphics depicting the relationships admissible Visual
Damage-
energy (VDADL¨energy) and minimum Visual Damage-energy (VDm,n-energy),
respectively.
CA 03031038 2019-01-16
8
In figure 9b, the curve A represents the case when minimum Visual Damage
(VDm,n) is
monotonically increasing with the energy, while curve B represents the case
when it is not.
Figure 10 shows three graphics depicting minimum Visual Damage-admissible
Visual
Damage (VDmin-VDADL) relationship. If the minimum Visual Damage (VDmin)
monotonically
increases with the energy level, case A is applicable; if the minimum Visual
Damage
(VDm,n) does not monotonically increase with the energy, case B1 and B2 are
applicable.
Figure 11 shows a graphic depicting the relationship Structural Damage-Visual
Damage
(SD-VD) found using the method called "Energy-Dependent Approach".
DETAILED DESCRIPTION OF THE INVENTION
Depending on the maximum lightning strike energy level expected in a defined
aircraft zone
and its relation to the admissible Structural Damage (ADL) applicable to that
zone, two
different embodiments for calculating the area of the Visual Damage are given
(the
Bounding-box approach and the Energy-dependent approach).
Method 1: Boundinq-box approach
As previously stated, this embodiment is valid for cases when the admissible
Structural
Damage (ADL) of the aircraft zone in which the mark is located is above the
Structural
Damage (SD) associated to the maximum lightning strike energy level or peak
current
expected on the aircraft section studied, see Figure la.
In this method a strong hypothesis is done: for the aforementioned maximum
energy level, it
is assumed that there is no relationship between the Visual Damage (VD) and
the Structural
Damage (SD), which leads to analyze Visual Damage (VD) and Structural Damage
(SD) as
statistically independent parameters. This hypothesis lies down on the
assumption that, for
an adequate number of specimens, as Visual Damage (VD) and Structural Damage
(SD)
have a monotonically increasing dependence with the energy and as the area
under study is
associated to the maximum foreseen energy, higher damages than the limits
calculated are
not expected.
The method comprises the following steps:
CA 03031038 2019-01-16
9
1) Sort the data by the primary parameter (i.e. energy level or peak current).
The highest
energy level tested corresponds to the most energetic lightning strike the
aircraft skin
zone in which the mark is located is going to experience.
2) Calculate the confidence interval of each type of damage for the highest
energy level,
taking into account the amount of specimens available for the study and the
confidence level that wants to be achieved for each type of damage (VD, SD)
independently. The calculated limits define a "bounding-box" where the data is
confined, see Figure 3.
3) The maximum Visual Damage (VDmax) determined by the confidence interval is
the
admissible Visual Damage (VDADL).
To verify said calculation, it must be ensured that the admissible structural
damage (ADL) of
the mark is higher or equal than the upper limit of the bounding box, i.e.,
than the maximum
Structural Damage (SDmax). Hence, the admissible Visual Damage (VDADL)
corresponds to
the rightermost limit of the box (VDmax), see Figure 4.
Method 2: Energy-dependent approach
If the admissible Structural Damage (ADL) target is not associated to the
maximum expected
energy in the studied aircraft zone, see Figure lb, a second method is used to
determine the
admissible Visual Damage (VDADL).
With this method, it is possible to determine and validate the final
admissible Visual Damage
(VDADL) figure while minimizing the number of tests. Lightning strike testing
is expensive; this
approach allows:
(a) estimating the energy level that is in the acceptable range of Visual
Damage (VD)
for a defined admissible structural damage (ADL), thus reducing as much as
possible the
number of tests.
(b) Once the energies that must be tested are obtained, the admissible Visual
Damage (VDADL) value can be determined.
CA 03031038 2019-01-16
0
As aforementioned, Visual Damage (VD) and Structural Damage (SD) follow in
average a
monotonically increasing dependence with the energy. From this it can be
deducted that
Visual Damage (VD) and Structural Damage (SD) also follow in average a
monotonically
increasing dependence, as the empirical data verify.
The idea behind this embodiment is to appropriately model this Structural
Damage-Visual
Damage (SD-VD) relationship across all different energy levels, as opposed to
the first case,
where the Visual Damage (VD) and Structural Damage (SD) were analyzed
independently
for the maximum energy level expected. A prediction confidence interval of the
Structural
Damage-Visual Damage (SD-VD) relationship is calculated.
When analyzing the Structural Damage (SD) and Visual Damage (VD) parameters
independently (embodiment 1) the independent predicted intervals are higher
because not all
the available information is used. Taking into account that Structural Damage
(SD) and
Visual Damage (VD) follow a monotonically increasing relationship allows
discarding some
areas (VD, SD); hence, the prediction interval is more precise. In Figure 5, a
comparison of
qualitative results when applying the bounding-box method and the energy-
dependent
method reveals a shaded area that corresponds to the area that can be
discarded when
using the energy-dependent method, which represents better the underlying
physics.
In order to achieve the first goal (i.e., estimating the energy that must be
tested for being in
an acceptable range of Visual Damage (VD) and Structural Damage (SD)), the
following
issues need to be solved:
= Previous empirical data demonstrate that the Visual Damage-Structural
Damage (VD-
SD) relationship is clearly exponential, see Figure 6; as the energy
increases,
Structural Damage (SD) increases exponentially as well as its dispersion. It
would be
possible to accurately define this behavior with an extensive lightning strike
test
campaign, but this option is not feasible as it would imply a considerable
financial
impact.
= To overcome this issue, a mathematical model needs to be used to
determine a
prediction interval for the Structural Damage (SD). Unfortunately, there is no
existing
model able to adapt to an exponential curve with exponential increase of
dispersion;
CA 03031038 2019-01-16
11
standard models are driven by the Structural Damage (SD) scatter of the
maximum
energy level included in the calculations. As a result, extremely high values
of
Structural Damage (SD) are predicted for the lower energy levels, giving
useless
results from an applied technological perspective. This is clearly a fault of
the
mathematical model as empirical data show that the limit predicted by the
standard
models is excessively conservative, see Figure 6.
= As this method allows making estimations for lower energies, it must be
assured that
higher energy levels will not endanger the established allowable Visual Damage
(VD):
Structural Damages (SD) higher than the admissible structural damage (ADL)
cannot
be permitted for Visual Damages (VD) lower than the admissible Visual Damage
(VDADL), see Figure 7b, as this situation could lead to a safety issue.
To solve the aforementioned issues it is necessary to understand two
relationships: the
relationship between the Visual Damage (VD), the Structural Damage (SD) and
the
admissible structural damage (ADL); and the relationship between the Visual
Damage (VD)
and the energy level.
= Visual Damage-Structural Damage (VD-SD) relationship with the admissible
structural damage (ADL):
a. When calculating prediction intervals, due to the exponential increase of
the
scatter with the energy (see Figure 6), when taking into account higher energy
levels, the prediction interval is dominated by the Structural Damage-Visual
Damage (SD-VD) values of the highest energy level. See Figure 7a; when
adding higher energy levels to the calculations (ea - e4+e3 ea+ e3+e2
e4+e3+e2+el; being e, different energy levels where e4<e3<e2<el), the
Structural Damage (SD) prediction curve (the maximum values of Structural
Damage (SD) predicted for a Visual Damage (VD) range) is modified as it is
strongly influenced by the higher energy level.
b. See Figure 7a. For the ei case, the prediction interval never intersects
the
admissible structural damage (ADL) line, hence, statistically, for any value
of
Visual Damage (VD), the Structural Damage (SD) obtained will be higher than
the admissible structural damage (ADL).
c. See Figure 7a. For the e3 case (only energy levels e3 and ea are taken into
account) the prediction interval decreases as the scatter of damages
associated to e3 is lower than the one associated to el. Hence, the prediction
interval intersects the admissible structural damage (ADL) line, allowing to
CA 03031038 2019-01-16
12
define an admissible Visual Damage (VDADL3) that assures that for lower
values of Visual Damages (VD), the Structural Damage (SD) will be lower
than the admissible Structural Damage (ADL).
d. Consequently, the highest energy level taken into account for the
statistical
method will determine the admissible Visual Damage (VDADL). As the highest
energy level decreases, the admissible Visual Damage (VDADL) will increase
(i.e. VDADL1 <VDADL2<VDADL3<VDADL4).
e. Therefore, different admissible Visual Damage (VDADL), which fulfil de
condition of Structural Damage (SD) < admissible Structural Damage (ADL),
can be defined for different energy levels.
= Visual Damage (VD) relationship with the energy level:
a. Although there are different possible admissible Visual Damages (VDADL) for
different energy levels that fulfil the condition of Structural Damage (SD) <
admissible Structural Damage (ADL), not all values are viable.
b. As already mentioned, each energy level has a Visual Damage (VD) scatter
associated to it (see Bounding-box Approach); hence a range of Visual
Damage (VD) [VDm,n, VDmad can be predicted for each energy level.
c. Hence, each energy level has an admissible Visual Damage (VDADL) and
minimum Visual Damage (VDmin) associated to it. See Figure 8.
d. The admissible Visual Damage (VDADL) corresponding to a specific energy
level can be greater than the minimum Visual Damage (VDm,n) of higher
energy levels, which implies that Structural Damage (SD) higher than the
admissible Structural Damage (ADL) could be found for Visual Damage (VD)
lower than admissible Visual Damage (VDADL). See Figure 7b. Contrary to
admissible Visual Damage (VDADL), minimum Visual Damage (VDm,n) does not
always follow a monotonically increasing relationship with the energy (see
Figure 9), reason why this circumstance can occur.
Therefore, taking into account (a) the relationship between the Visual Damage-
Structural
Damage (VD-SD) with the admissible structural damage (ADL) and (b) the Visual
Damage
(VD) relationship with the energy, the following condition shall be met:
trDAnz KI/Drun'c IrDmin e > [1]
CA 03031038 2019-01-16
13
being e* the highest energy level included in the calculations and VDADL* its
associated
admissible visual damage. This condition assures that there will be no
Structural Damages
(SD) higher than the admissible Structural Damage (ADL) for Visual Damages
(VD) lower
than the admissible Structural Damage* (VDADL.).
= It is possible to set the relationship between admissible Visual Damage
(VDADL) and
the energy level e, as well as between the minimum Visual Damage (VDmin) and
the
energy level e, see Figure 9. The admissible Visual Damage (VDADL) follows a
monotonically decreasing function. The minimum Visual Damage (VDmin) follows a
monotonically increasing relationship with the energy e, up to an energy level
where,
depending on several parameters, it may or may not continue following a
monotonically increasing relationship.
= Making use of the aforementioned knowledge, a relationship between the
minimum
Visual Damage (VDm,r,) and the admissible Visual Damage (VDADL) can be
established, which allows to determine the optimal admissible Visual Damage
(VDADL), i.e., the maximum admissible Visual Damage (VDADL) that fulfills
condition
[1], see Figure 10. Going back to the relationship maximum admissible Visual
Damage-energy (VDADL-e), the maximum energy level at which it should be tested
can be determined.
Therefore, to determine the energies at which it should be tested, these steps
need to be
followed:
1) Calculate the admissible Visual Damage (VDADL) and minimum Visual Damage
(VDmin) estimates for different energy levels using data from previous test
campaigns/real damages.
2) Find out the relationship between the admissible Visual Damage (VDADL) and
the
energy levels, as well as between the minimum Visual Damage (VDmin) and the
energy levels.
3) Relate admissible Visual Damage (VDADL) and minimum Visual Damage (VDmin)
to
find the optimal admissible Visual Damage
(VDADL*),
VDADL < V V D rtlin e e*.
CA 03031038 2019-01-16
14
4) Find out the energy level that corresponds to the optimal admissible Visual
Damage
(VDADL*) using the relationship found in step 2.
Once the maximum energy to test is obtained, the next steps need to be
followed to
empirically validate the estimated admissible Visual Damage (VDADL):
1) Obtain data by testing at least two, recommended three or more, close
energy levels.
The higher the number of specimens, the smaller prediction intervals.
2) Find the best suitable regression to relate the visual damage (VD) and the
structural
damage (SD), being the visual damage (VD) the independent variable and the
structural damage (SD) the dependent one (SD = f(VD)). This regression can be
linear or non-linear, depending on the specific case (e.g., depending on the
material
and lightning surface protection).
3) The prediction interval is calculated taking into account the number of
specimens, the
degree of confidence, the quality of the adjustment and the data dispersion by
using
existing mathematical methods found in the literature.
4) Determine the allowable visual damage (VDADL), which corresponds to the
intersection between the admissible Structural Damage (ADL) and the structural
damage (SD) prediction interval. This intersection defines a safe area where
it is
assured that no structural damage (SD) higher than the admissible Structural
Damage (ADL) will be encountered for a Visual Damage (VD) lower than the
admissible Visual Damage (VDADL), see Figure 11.